diff --git a/sources/platform/images/apify_mcp_server.png b/sources/platform/images/apify_mcp_server.png index 5dbf87144e..f0642e94ae 100644 Binary files a/sources/platform/images/apify_mcp_server.png and b/sources/platform/images/apify_mcp_server.png differ diff --git a/sources/platform/integrations/ai/mastra.md b/sources/platform/integrations/ai/mastra.md index 9259f33ed8..e042feac21 100644 --- a/sources/platform/integrations/ai/mastra.md +++ b/sources/platform/integrations/ai/mastra.md @@ -66,7 +66,7 @@ Instantiate the Mastra MCP client: const mcpClient = new MastraMCPClient({ name: 'apify-client', server: { - url: new URL('https://actors-mcp-server.apify.actor/sse'), + url: new URL('https://mcp.apify.com/sse'), requestInit: { headers: { Authorization: `Bearer ${process.env.APIFY_TOKEN}` } }, @@ -122,17 +122,10 @@ console.log(response.text); await mcpClient.disconnect(); ``` -Before running the agent, we need to start the [Actors MCP Server](https://apify.com/apify/actors-mcp-server) by sending a request: - -```bash -curl https://actors-mcp-server.apify.actor/?token=YOUR_APIFY_TOKEN&actors=apify/rag-web-browser,clockworks/free-tiktok-scraper -``` - -Replace `YOUR_APIFY_TOKEN` with your Apify API token. You can also open the URL in a browser to start the server. - :::note Use any Apify Actor -Since it uses the [Actors MCP Server](https://apify.com/apify/actors-mcp-server), swap in any Apify Actor from the [Apify Store](https://apify.com/store) by updating the startup request’s `actors` parameter. No other changes are needed in the agent code. +Since it uses the [Apify MCP Server](https://mcp.apify.com), swap in any Apify Actor from the [Apify Store](https://apify.com/store) by updating the startup request’s `actors` parameter. +No other changes are needed in the agent code. ::: @@ -179,7 +172,7 @@ process.env.OPENAI_API_KEY = "your-openai-api-key"; const mcpClient = new MastraMCPClient({ name: 'apify-client', server: { - url: new URL('https://actors-mcp-server.apify.actor/sse'), + url: new URL('https://mcp.apify.com/sse'), requestInit: { headers: { Authorization: `Bearer ${process.env.APIFY_TOKEN}` } }, @@ -223,7 +216,8 @@ await mcpClient.disconnect(); - [Apify Actors](https://docs.apify.com/platform/actors) - [Mastra Documentation](https://mastra.ai/docs) -- [Apify MCP Server](https://apify.com/apify/actors-mcp-server) +- [Apify MCP Server](https://mcp.apify.com) +- [How to use MCP with Apify Actors](https://blog.apify.com/how-to-use-mcp/) - [Apify Store](https://apify.com/store) - [What are AI Agents?](https://blog.apify.com/what-are-ai-agents/) - [How to Build an AI Agent](https://blog.apify.com/how-to-build-an-ai-agent/) diff --git a/sources/platform/integrations/ai/mcp.md b/sources/platform/integrations/ai/mcp.md index 2feefeac80..9458e8dfe8 100644 --- a/sources/platform/integrations/ai/mcp.md +++ b/sources/platform/integrations/ai/mcp.md @@ -17,8 +17,8 @@ The _Apify Model Context Protocol (MCP) Server_ allows AI applications to connec You can use the Apify MCP Server in two ways: - _HTTPS Endpoint_ `mcp.apify.com`: Connect your MCP client through OAuth or by including `Authorization: Bearer ` header in your requests. - - `https://mcp.apify.com` for streamable transport - - `https://mcp.apify.com/sse` for legacy SSE transport + - `https://mcp.apify.com` for streamable transport (recommended) + - `https://mcp.apify.com/sse` for SSE transport (legacy) - _Standard Input/Output (stdio)_: Ideal for local integrations and command-line tools such as the Claude for Desktop client. - Set MCP client server command to `npx @apify/actors-mcp-server` and environment variable `APIFY_TOKEN` to your Apify API token - See `npx @apify/actors-mcp-server --help` for more options @@ -108,9 +108,9 @@ In the client settings, you need to provide server configuration: By default, the main Actors MCP Server starts with a single default [RAG Web Browser Actor](https://apify.com/apify/rag-web-browser). However, you can fully customize which Actors are available: -- _Dynamic adding during a session:_ If your client supports it, the agent itself can add Actors dynamically by name (using the `add-actor` operation) at runtime. For example, after using `search-actors` to find an Actor’s name, calling `add-actor` with that name will load it. - _Tools for adding and removing Actors are enabled by default._ - You can disable these tools by setting the parameter `?enableAddingActors=false` in the MCP Server URL, or with the CLI flag `--enable-adding-actors=false` (can also be set in Claude for Desktop config args as `--enable-adding-actors=false`). +- _Dynamic adding during a session:_ If your client supports it, the agent itself can add Actors dynamically by name (using the `add-actor` operation) at runtime. For example, after using `search-actors` to find an Actor’s name, calling `add-actor` with that name will load it. + _Tools for adding and removing Actors are enabled by default._ + You can disable these tools by setting the parameter `?enableAddingActors=false` in the MCP Server URL, or with the CLI flag `--enable-adding-actors=false` (can also be set in Claude for Desktop config args as `--enable-adding-actors=false`). Not all MCP client frameworks allow dynamic tool addition at runtime, but Apify’s own tester client does, if adding Actors is enabled. - _Via url:_ If you are using Streamable HTTP or SSE protocol, you could add `actors` query parameter with Actor names separated by comma: diff --git a/sources/platform/integrations/index.mdx b/sources/platform/integrations/index.mdx index 9ff46ac7af..f64ca0de1c 100644 --- a/sources/platform/integrations/index.mdx +++ b/sources/platform/integrations/index.mdx @@ -146,15 +146,10 @@ The Apify platform integrates with popular ETL and data pipeline services, enabl
-If you are working on an AI/LLM-related project, we recommend you look into the Langchain integration. +If you are working on AI/LLM-related applications, we recommend looking into the many integrations with popular AI/LLM ecosystems. +These integrations allow you to use Apify Actors as tools and data sources. - + + + +